BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Phylogenetic Inference of Multidomain Evolution - Dannie Durand (C
 arnegie Mellon University\, Pittsburgh)
DTSTART:20110620T150000Z
DTEND:20110620T160000Z
UID:TALK30198@talks.cam.ac.uk
CONTACT:Florian Markowetz
DESCRIPTION:Multidomain proteins are characterized by a mosaic of sequence
 \nfragments that encode structural or functional modules\, called\ndomains
 .  Multidomain families evolve via domain shuffling: insertion\,\nduplicat
 ion\, and deletion of domains\, as well as fusion and fission of\nadjacent
  coding sequences.  Multidomain families play a central role\nin cell sign
 aling and adhesion.  Preferential expansions in the size\nand complexity o
 f multidomain families are associated with key\nevolutionary transitions\,
  including the emergence of the Metazoa and\nthe evolution of vertebrates.
   As a consequence of their functional\nrepertoire\, multidomain families 
 are of central importance in immune\nresponse\, tissue repair\, neural dev
 elopment\, and cancer.  The\nmodularity of multidomain families is linked 
 to their functional roles\nin cell signaling and adhesion and was a drivin
 g force in the\nevolution of multicellularity in animals.  It has been pro
 posed that\nmodularity confers functional plasticity: changes in modular\n
 structure\, through acquisition\, loss\, or replacement of a domain\,\npot
 entially result in rapid rewiring of cellular networks.\n\nUnderstanding t
 he connection between protein evolution and network\nevolution requires me
 thods to reconstruct multidomain evolution.\nHere\, I present algorithms t
 o infer the history of domain\nduplications\, losses\, and insertions in a
  multidomain family\, as well\nas the timing of these events and the ances
 tral domain architectures.\nKey features of my approach include explicit r
 epresentation of domain\nshuffling events and a model that captures sequen
 ce variation within\ndomain families.  I demonstrate the utility of my met
 hods with\nin-depth analyses of well-studied multidomain families\, as wel
 l as a\ngenome-scale analysis of domain families in human.  My results sug
 gest\nthat a remarkably greater amount of domain shuffling may have occurr
 ed\nthan predicted by simpler models that do not consider sequence\nvariat
 ion.  Further\, they underscore the importance of accurate\nevolutionary r
 econstruction for homology-based function prediction and\nevolutionary sys
 tems biology.\n\nWebsites \n* http://www.cmu.edu/bio/faculty/durand.html\n
 * http://www-2.cs.cmu.edu/~durand/Lab/
LOCATION:Cancer Research UK Cambridge Research Institute\, Lecture Theatre
END:VEVENT
END:VCALENDAR